PSI - Issue 45

Teresa Magoga et al. / Procedia Structural Integrity 45 (2023) 28–35 Author name / Structural Integrity Procedia 00 (2019) 000 – 000

29

2

1. Introduction Fatigue is an important factor in the structural design and through-life management of weight-optimised ships. Such ships are typically constructed from marine-grade aluminium alloys or high tensile steel. The reduction in the scantlings due to the use of these materials has in some cases lead to fatigue cracking.

Nomenclature D

Fatigue damage

FEA FL FL d FL m FL p

Finite Element Analysis

Fatigue Life

[years] [years] [years] [years]

Design fatigue life

Time from commissioning to discovery of first crack

Predicted fatigue life Stress concentration factor

K

m 1 m 2

Inverse slope of S-N curve f or N ≤ 5x10 Inverse slope of S-N curve for 5x10

6 cycles

6 < N ≤ 10 8 cycles

n

Number of stress cycles

N

Number of stress cycles to failure

S-N

Fatigue resistance (stress range versus no. cycles to failure)

 , S Stress range

[MPa] [MPa] [MPa] [MPa] [MPa]

Reference stress range of fatigue strength at 2x10

6 cycles

 c

 max Maximum stress range

 crest Stress under design crest landing condition  hollow Stress under design hollow landing condition

Number of stress ranges identified in a time history



Methods to assess the fatigue strength of welded joints include the nominal stress approach, hot-spot stress approach, effective notch stress approach, fracture mechanics, and component testing. The first three methods are categories of the S-N curve concept, which is based on empirical data collected from fatigue tests of common structural details. If the detail of interest is represented in a standard then use of the nominal stress approach is valid. However, the selection of a suitable reference detail, and assessment of structures characterised by complex geometry and/or load combinations, can be difficult (Blagojevic et al. 2002, Collette and Incecik 2006, Shen et al. 2016, Soliman et al. 2015, Tveiten et al. 2007). As such, following an industry recognised code to conduct fatigue analysis can require interpretation by the analyst. For example, the aluminium structural design code Eurocode 9 (Technical Committee CEN/TC 250 1999) uses S-N curves based on nominal stresses. If the crack initiation site is a weld toe and the nominal stresses in the joint are not clearly defined, the hot-spot stress approach is preferable. However, hot-spot S-N curves must be available. In addition, selection of an applicable detail or S-N curve from a fatigue design code for a complex structure can be challenging and somewhat subjective (Sielski 2008). Such complexities are reflected in the variability of outcomes in the literature. Al Zamzami and Susmel (2017) conducted a comparative assessment of different approaches based on extensive experimental data, finding that use of industry design curves with the nominal stress approach provides an adequate level of accuracy. In the fatigue analysis of trapezoidal joints in a fast ferry, Garbatov et al. (2010) assumed t hat the neighbouring ‘low - gradient’ is the nominal stress but did not provide a quantitative definition of this stress. In comparison, Soliman et al. (2015) conducted fatigue analysis of a fillet welded detail in an aluminium catamaran using the hot-spot stress approach because many of the construction details have no direct match in design guides. Further, there are numerous variants in modelling and procedures for both the nominal stress and hot-spot stress concepts (Radaj et al. 2009). Tveiten et al. (2007) demonstrated the variability in the stress levels at highly stressed locations with different Finite Element Analysis (FEA) models and stress extrapolation techniques. Information relating to fatigue failures and operational profile is valuable for the design, acquisition, and management of ships, and to help validate fatigue evaluation approaches. The International Association of